Fritz Haber - Nobel Lecture

RITZ

ABER

The synthesis of ammonia from its elements

Nobel Lecture, June 2,

1920

The Swedish Academy of Sciences has seen fit, by awarding the Nobel
Prize, to honour the method of producing ammonia from nitrogen and hy-
drogen. This outstanding distinction puts upon me the obligation of explain-

ing the position occupied by this reaction within the subject of chemistry as
a whole, and to outline the road which led to it.

We are concerned with a chemical phenomenon of the simplest possible

kind. Gaseous nitrogen combines with gaseous hydrogen in simple quantita-

tive proportions to produce gaseous ammonia. The three substances involved
have been well known to the chemist for over a hundred years. During the
second half of the last century each of them has been studied hundreds of
times in its behaviour under various conditions during a period in which a

flood of new chemical knowledge became available. If it has not been until
the present century that the production of ammonia from the elements has
been discovered, this is due to the fact that very special equipment must be
used and strict conditions must be adhered to if one is to succeed in obtaining
spontaneous combination of nitrogen and hydrogen on a substantial scale,
and that a combination of experimental success with thermodynamic con-

siderations was needed.

It was particularly significant that earlier attempts had not succeeded, even

fleetingly, in achieving with absolute certainty a spontaneous union of nitro-

gen and hydrogen to form ammonia. This gave rise to the prejudice that such
a production of ammonia was impossible, and in the course of time this en-

joyed considerable support in chemical circles. Such prejudice leads one to

expect pitfalls which, far more than clearly-defined difficulties, deter one
from becoming too deeply involved in the subject.

A narrow professional interest in the preparation of ammonia from the

elements was based on the achievement of a simple result by means of special

equipment. A more widespread interest was due to the fact that the synthesis
of ammonia from its elements, if carried out on a large scale, would be a
useful, at present perhaps the most useful, way of satisfying important na-
tional economic needs. Such practical uses were not the principal purpose of

A M M O N I A S Y N T H E S I S F R O M E L E M E N T S

327

my investigations. I was never in doubt that my laboratory work would

produce no more than a scientific confirmation of basic principles and a cri-

terion of experimental aids, and that much would need to be added to any
success of mine to ensure economic success on an industrial scale. On the
other hand I would hardly have concentrated so much on this problem had
I not been convinced of the economic necessity of chemical progress in this
field, and had I not shared to the full Fichte’s conviction that while the imme-
diate object of science lies in its own development, its ultimate aim must be
bound up in the moulding influence which it exerts at the right time upon

life in general and the whole human arrangement of things around us.

Since the middle of the last century it has become known that a supply of

nitrogen is a basic necessity for the development of food crops; it was also

recognized, however, that plants cannot absorb the elementary nitrogen
which is the main constituent of the atmosphere, but need the nitrogen to
be combined with oxygen in the form of nitrate in order to be able to assim-

ilate it. This combination with oxygen can start with combination with
hydrogen to form ammonia since ammonium nitrogen changes to saltpetre

nitrogen in the soil.

Under natural conditions the soil does not lose its fixed nitrogen. Green

plants use it to synthesize complicated substances without changing it into
elementary nitrogen. Animals and humans ingest it with the plants and re-
turn it to the soil in fixed form in their excretions and finally with their de-
ceased remains. Putrefaction and combustion does destroy a certain amount of
fixed nitrogen, but Nature makes good the loss when, during thunderstorms,
lightning combines nitrogen and oxygen in the upper layers of the atmo-
sphere, which is then washed down by the rain. To this nitrogen-fixing ac-
tion of electrical discharge as a source of bound nitrogen is added the activity
of soil bacteria, some of which live free while others are to be found settled
in the root nodules of many plants, converting free nitrogen into bound ni-
trogen.

Agricultural husbandry essentially maintains the balance of bound nitro-

gen. However, with the advent of the industrial age, the products of the soil
are carried off from where the crops are grown to far-off places where they
are consumed, with the result that the bound nitrogen is no longer returned
to the earth from which it was taken.

This has caused the world-wide economic necessity of supplying bound

nitrogen to the soil. This need is increased by national economic considera-
tions, which, with the denser population of industrialized countries, call for

328

1 9 1 8 F . H A B E R

increased agricultural productivity at home, and it is yet further increased by

the fact that expanding industry requires fixed nitrogen for many of its own
chemical purposes. The demand for nitrogen, like that for coal, indicates how
far removed our way of life has become from that of the people who ((them-
selves do fertilize the soil they cultivate)).

Agriculture, always the main consumer, is not satisfied with a supply of

nitrogen alone - potash and phosphates are equally indispensable - but the
world possesses far fewer natural resources for meeting nitrogen require-
ments. And so, naturally, concern over nitrogen supplies has become the
first of the great obstacles that lie along the highway of world commerce
upon which we have been travelling in recent decades.

Our way of thinking, so used to interpreting historical events in the con-

text of man’s unchangeable nature, easily misleads us into overlooking the
enormous turning-point in the history of mankind represented by the last
hundred years. In earlier periods the need for energy was satisfied by men’s
physical labour and by the use of wind and sun, which are older than our-

selves and will outlive our life conditions. The past century has opened the
floodgates for the energy stored in coal, and has introduced ways of life in
industrialized countries in which the physical labour of men merely operates
a relay to release the hundred times more powerful energy of coal into the
lifestream of international commerce. Technical needs have arisen for which
we only too easily find ourselves unprepared through a lack of adequate
scientific development. The present state of affairs in the world, with the
after-effects of the War in Central Europe placing an overwhelming load on
our scientific work, makes this only too plain.

The need for opening up new sources of nitrogen became clearly apparent

at the turn of the century. Since the middle of the last century we had been
drawing upon the supply of saltpetre nitrogen which Nature had deposited
in the high-mountain deserts of Chile. By comparing the fast-rising require-
ments with the calculated extent of these deposits it became clear that to-

wards the middle of the present century a major emergency would be una-
voidable, unless the chemistry found a way out.

The problem was not a new one to the chemists. When they began to

distil coal they came across ammonia among the distillation products and

this, in the form of ammonium sulphate, found application in agriculture.
While in 1870 ammonia was a tiresome by-product of the gas industry, by

1900

it had become a very valued companion to combustible gases and the

coke industry was in full swing everywhere to adapt furnaces to its by-pro-

A M M O N I A S Y N T H E S I S F R O M E L E M E N T S

329

duction. Its origin from the fixed nitrogen of coal was understood; an im-

provement in its yield, which by the normal process was hardly more than

1/5 of the nitrogen content in the coal, had been widely studied. But no

satisfactory solution seemed likely in that direction.

With an average content of about 1% of fixed nitrogen, coal could not be

processed for obtaining nitrogen only. The delivery of nitrogen as a by-

product set limits to its production which made it impossible to make good a
future deficiency of saltpetre from this source. It was clear that the demand

for fixed nitrogen, which at the beginning of this century could be satisfied
with a few hundred thousand tons a year, must increase to millions of tons.
A demand of this order could only be met from one source-from the im-
mense supply of elementary nitrogen available in our atmosphere-and the

binding would have to be achieved by chemical means to the simplest and
most widely available chemical elements, if the solution was to measure up
to the demand. Just as the raw-material situation of our Earth indicates ele-

mentary nitrogen as the starting material, so ammonia or nitric acid are in-

dicated as end products by the requirements of plants. The task thus became

the combining of elementary nitrogen with oxygen or water.

This again was not a new or untried chemical problem. The combining of

nitrogen with hydrogen to form ammonia as with oxygen to produce nitr-

ates had already occupied science and, to some extent, technology.

Combination with hydrogen directly from the elements had been induced

by various forms of electrical discharge, which of course resulted in an ener-
gy consumption of alarming proportions. Indirect combination, on the
other hand, had been developed with remarkable technical results; the ni-

trogen was combined with other elements and this combination was then
hydrolysed with water whereby ammonia was splitt off. Only the spon-
taneous association of the elements was unknown when, in 1904, I began to
occupy myself with the subject; it was held to be impossible after pressure,
heat, and the catalytic action of platinum sponge had been found unable to

produce the effect.

The indirect method has occupied the attention of scientists and technol-

ogists ever since Margueritte and Sourdeval, basing themselves on earlier
work by Bunsen and Playfair, developed it to the stage of sample produc-
tion in 1860. Caustic baryte and coal at high temperatures with nitrogen
yielded barium cyanide. At lower temperatures this combination broke
down in the presence of water vapour, yielding ammonia and creating ba-
rium hydroxide which returned to the process. Thus, during alternate for-

330

1918 F.HABER

mation and breaking down of barium cyanide, a continuous yield of am-
monia and carbon dioxide was obtained from coal, water and elementary
nitrogen. In the half-century following the publication by Margueritte and

Sourdeval, this indirect method, the early technical execution of which

made excessive demands on the reaction vessels, has been studied afresh in
many modified forms.

Barytes could be replaced by heat-resistant oxides of other metals or semi-

metals. The process of nitrogen fixation could be broken down into partial
steps, first forming, by reduction, the metal, semi-metal or metal carbide
which would, in a subsequent reaction, take up the nitrogen. As a solution
to the problem of ammonia synthesis the result has never been entirely satis-
factory.

If the reduction of oxide and the fixation of nitrogen took place in a single

process then this required an extremely high temperature. If the process were
split up, intermediate products were obtained which reacted more easily
with nitrogen. But the intermediate product-metal, semi-metal, or carbide

- then demanded, for its own production from the massive reserves of nat-

ural products, precisely those conditions which led to an uneconomical
consumption of electrical energy, either by electrolytic or electrothermal
means.

The more tightly knit nitrogen molecule does not break down as easily

as oxygen, the next element in the periodic system. The abundant examples

we have of autoxidation are thus matched by a complete lack of spontaneous
reaction of elementary nitrogen in the inanimate world at normal tempera-
tures. The inaccessibility of nitrogen nullified all the many efforts made to
develop a technical ammonia process.

In only one respect has the study of indirect methods of synthesizing am-

monia from the elements been able to get round the difficulties. Frank and
Caro obtained the important calcium cyanamide through the action of ni-
trogen on calcium carbide obtained from lime and coal in the electric arc.

Splitting the calcium cyanamide with water produces ammonia, and the

process takes place in the soil without any particular help from us, once the
cyanamide has been added to the soil as fertilizer. The saving in factory pro-
cessing achieved by this, plus the fact that the only raw materials required
are lime, coal and nitrogen, have been important factors in the establishment
of the process.

Efforts to combine nitrogen with oxygen go back further than those aimed

at combining it with hydrogen. The basic fact of the combination of ni-

A M M O N I A S Y N T H E S I S F R O M E L E M E N T S

331

trogen with oxygen during sparking had already been observed by Caven-
dish and Priestley. In this case the first product is nitric oxide, which converts
to nitric acid in a spontaneous reaction with oxygen and water. The nitric
oxide synthesis is a process requiring heat, and unless energy is supplied can,
for thermodynamic reasons, only occur spontaneously to any appreciable
extent at extremely high temperatures. However, the supply of energy re-
quired at normal temperatures is so small that disadvantage of having to pro-
vide it is outweighed by the advantage of needing only air and water as raw
materials. No better and more economical process for the binding of nitro-
gen could therefore be devised if some means could be found for converting
electrical energy into this kind of chemical energy without waste.

The example of Nature, which produces the reaction via lightning and

Cavendish’s earlier successful imitation of this with electric sparks, coupled
with the outstanding electrotechnical developments of the final decades of

previous century, increasingly brought this method of solving the nitrogen
problem to the fore, as professional circles became less and less satisfied with

the progress achieved through combining nitrogen with hydrogen. The
brilliant developments which these efforts produced in the early years of
this century are general knowledge. The main outlines of the technical de-
sign coupled particularly with the names of Birkeland and Eyde, of Schoen-
herr and of Pauling, have for years been the object of a great deal of interest

among experts.

Installations on a considerable scale were built in a number of places and

the method was evidently well suited to making use of the vast supply of
energy which could be derived from waterfalls for chemical purposes; but
this method of synthesizing nitrogen has still not reached the levels of pro-
duction which it appeared to promise. Its progress is limited by the fact that
with a consumption of one kilowatt-hour no more than 16 grams of nitro-
gen are converted into nitric acid, whilst a complete conversion of electrical
to chemical energy ought to yield 30 times as much. An explanation of this
has been given by Muthmann and Hofer, who have demonstrated that the
high-tension arc used in this process, acts as a Deville’s heat evaporation

chamber.

The formation of nitric oxide is determined, and limited, by thermal con-

ditions in the arc and its surroundings. Determination of the thermodynamic
equilibrium of nitric oxide synthesis by Nernst confirmed this explanation.
An extrapolatron of his experimental results and the best figures for the spe-
cific heat of the gases involved up to the temperature of 3,000

C or 4,000

332

1918 F.HABER

led to the remarkable conclusion that more than

1½

times or twice the tech-

nical yield per kilowatt-hour could still not be achieved when no re-forma-
tion of nitric oxide in the cooling circuit occurred at all. The source of the
low yield lay in the fact that the heating of a large air mass at very high tem-
peratures enabled only a small fraction to convert thermodynamically to
nitric oxide. In spite of the fact that, for a variety of reasons, this calculation

cannot pretend to considerable accuracy, its result obviously approaches the
truth. Practical experience has shown that no worthwhile saving of energy

can be achieved by heat regeneration, manifestly because the deterioration
of the quenching action involved militates against this. It is impossible to do
away with the arc discharge without deviating from the basic processes
which comply with the requirements of mass production.

However, it was perhaps not entirely impossible with a discharge arc to

get away from the temperature range in which rapid adjustment of the
thermodynamic balance covered every more favourable possibility of chang-
ing electrical into chemical energy. After all, the arc existed by virtue of the
constant production of units of higher energy in the form of gas ions caused
by the electrical energy of electronic impacts and it was not a priori evident
that the subsequent dissipation of energy in the form of heat precluded
everything else than the thermal result of nitric oxide synthesis, particularly
because Warburg and Leithaeuser had shown non-thermal synthesis of the
oxide by means of corona discharge.

This possibility aroused much interest during the first ten years of this cen-

tury and from 1907 led me to start investigations which I pursued over a
number of years. Development has so changed opinions during those short
ten years, that today it is already difficult to think oneself back into the views
then generally held ; yet it is indicative that so experienced and professional
a judge of chemico-technical possibilities as the "Badische Anilin- und Soda-
fabrik" thought so highly of my efforts to obtain improved efficiency from

electrical energy in the combining of nitrogen and oxygen, as to get in touch
with me in 1908 and - by providing their resources - to facilitate my work
on the subject; whereas they agreed with every caution to the proposal to
back me in the high-pressure synthesis of ammonia as well, approving it only
with hesitation.

In fact, even in my later judgement, the question of whether technical re-

search should be concentrated on the direct synthesis of ammonia from the
elements really depended on whether the consumption of energy during the
combining of nitrogen and oxygen could be considerably reduced. In tech-

A M M O N I A S Y N T H E S I S F R O M E L E M E N T S

333

nical questions, where the scales oscillate between success and failure, the
borderline between the two extremes is usually defined by modest differen-

ces in the consumption of energy and materials, and variations in these values

which lie within one decimal power will determine the result.

With a number of excellent assistants I therefore studied for some long

time the synthesis of nitric oxide by electrical discharge. I have searched
through the pressure range from 12 atm. to 25 mm mercury, cooled the arc
both from the wall and from the anode, and studied the relationship between
energy consumption and frequency up to about

50,000

cycles per second.

We obtained nitric oxide concentrations of 10% in air at decreased pressure
which indicated a deviation from the thermodynamic balance. Yields of
bound nitrogen were also noted for the same consumption in kilowatt-
hours which exceeded the earlier-mentioned value of 16 grams by 10-15%.
But in themselves these advantages were not conclusive, being moreover

achieved by methods which were hardly suited to adaptation to mass-pro-
duction. This series of investigations accordingly led to a strengthening of
the view that the technical solution was to be sought in the direct combina-

tion of nitrogen with hydrogen.

A study of nitric-oxide synthesis in pressure flames led to the same result.

It had been known since the days of Bunsen that the explosion of combustible
gas with nitrogen and oxygen gives rise to the formation of nitrous prod-
ucts, and Liveing and Dewar had described the formation of nitric acid in a
hydrogen flame under pressure. It appeared desirable to me also to familiar-
ize myself with this source of nitric oxide, in which heat was used as the
source of energy under conditions easily available in industry.

There were proposals to utilize the explosive reactions simultaneously in a

motor and as a source for the synthesis of nitric oxide. I myself placed no
faith in the linking of two such widely-differing functions. Yet the utiliza-

tion of the heat of flame gases appeared to me to be not incompatible with
the formation of nitric oxide, and worthy of closer investigation. This has
been extended over the flames of carbon monoxide, hydrogen and acetylene.
It was found that corresponding to

100

molecules of the main products of

combustion, carbon dioxide and hydrogen, 3 to 6 molecules of nitric acid
could be obtained. In the case of carbon monoxide and hydrogen this re-
quired increased pressure. Carbon monoxide had the advantage over the
hydrogenated gases, since the presence of water vapour in the hot products
of combustion favoured the reversion of the nitric oxide in the elements
along the cooling circuit. With this gas the molecular nitric oxide: carbon

334

1918 F.HABER

monoxide ratio could easily be brought, with air, to 3 :

100

and with a mix-

ture rich in oxygen to double that ratio. For technical utilization however,
these values were not sufficient incentive; the weight, which declined on the
direct combining of nitrogen with hydrogen, therefore again underwent an

increase.

I have not pursued further the combining of nitrogen and hydrogen by

corona discharge and by sparking. It seemed certain to me that this method

would not prove itself to be the most advantageous. In the final analysis the
assessment of each method rests upon the ratio between the energy consumed
and the yield, in other words, between coal consumption and nitrogen yield
(the consumption of hydraulic power being reckoned as the equivalent con-

sumption of coal).

Nothing seemed less hopeful, though, than the thought that the enforced

combining of nitrogen with hydrogen could be achieved with so little ener-
gy that one would have spare energy left over for the production of hydro-
gen. There remained merely the possibility of discovering the requirements
for spontaneous formation of ammonia from the elements. The positive heat
of formation of ammonia indicated that such a synthesis might be achieved
without the assistance of electrical energy. Against this there was the fact that
neither Deville nor Ramsay and Young had obtained ammonia by heating

nitrogen and hydrogen.

Ramsay and Young who, in 1884, during their study of the decomposi-

tion of the gas in the neighbourhood of 800

C had consistently observed a

trace of undecomposed ammonia, made great efforts to obtain this trace
from the elements at this temperature using iron as a carrier. But with pure
gases the experiment was unsuccessful. There was a point of uncertainty
here, and if this could be cleared up it would indicate the possibility of a
direct synthesis of ammonia from the elements.

therefore began tentatively to determine the approximate position of the

ammonia equilibrium in the vicinity of 1000

C. It now transpired that ear-

lier trials had only proved negative by accident; it was easy, in the vicinity
of 1000

C and using iron as a catalyst, to obtain the same ammonia content

from both approaches. The results of individual experiments fluctuated be-

tween 1/200% and 1/80%, and some discrepant values seemed to me to
point to the upper limit as the probable value; later more precise data proved

the lower limit to be the correct figure and showed the origin of the higher

values to be in the properties of the catalysts, which when fresh temporarily

bring about the synthesis of surplus ammonia.

A M M O N I A S Y N T H E S I S F R O M E L E M E N T S

335

It was further shown that the same results could be obtained with nickel as

with iron, and it was found that calcium and in particular manganese were
catalysts which would bring about a combination of the elements even at
lower temperatures. At 1,000

C the rate of reaction was adequate with a

small amount to produce continuously a comparatively large quantity of
ammonia. By having a circulation system which alternately brought the gas
at high temperature in contact with the metal and then washed out the am-
monia at normal temperature, the conversion of a given mass of gas to am-

monia could proceed stage by stage.

By determining results at a given pressure, temperature and initial mix-

ture of nitrogen and hydrogen, the state of the theory allowed obtainable
results to be approximately predicted for optional temperatures, pressures
and mixtures of nitrogen and hydrogen. In the light of the formula, it was
possible at once to foresee the increase of attainable maximum content with
decreasing temperature, its proportional relationship with the gas pressure,
and the fact that a mixture of 3 parts of hydrogen to

part of nitrogen must

result in the highest ammonia content.

The most important point realized at that time was that from the begin-

ning of red heat onwards no catalyst will produce more than a trace of am-
monia from the most favourable gas mixture at normal pressure, and that

even at greatly increased pressure the point of equilibrium must continue
very unfavourable. If one wished to obtain practical results with a catalyst

at normal pressure, then the temperature must not be allowed to rise much

beyond 300

At that point it seemed to me, in 1905, useless to pursue the problem

further. A combination of the elements had certainly been achieved, and the
requirements for large-scale synthesis had been outlined; but these require-

ments appeared so unfavourable that they deterred one from a deeper study
of the problem. The discovery of catalysts which would provide a rapid ad-

justment of the point of equilibrium in the vicinity of 300

and at normal

pressure seemed to me quite unlikely: and they have not been found any-
where in the 15 years that have since elapsed.

The synthesis of ammonia which had been demonstrated at normal pres-

sure could be carried out at high pressure on a laboratory scale without any
great difficulties. It needed only a slight modification of the pressure oven,

such as that used by Hempel 15 years earlier to carry out nitrogen absorption
in the case of indirect ammonia synthesis under pressures of up to 66 atmo-
spheres. But I did not think it worth the trouble; at that time I supported the

336

1918 F. HABER

widely-held opinion that a technical realization of a gas reaction at the be-
ginning of red heat under high pressure was impossible. Here the matter
rested for the next three years.

Already in 1906, on the other hand, a new determination of the ammonia

equilibrium proved necessary. In the course of his investigations into the

heat theorem which has been named after him, Nernst succeeded in finding

an approximate formula which permitted a prediction of the equilibria based
on the values of the heat effect and the so-called chemical constants. In the
case of ammonia this gave a deviation from the values obtained at my first

measurements which, as later became apparent, was caused by the original
value of the conventional chemical constant of hydrogen then used. This
deviation led to fresh determinations of the equilibrium which Nernst had

carried out at his Institute in a pressure oven indicated by him while I, in
collaboration with Robert le Rossignol, repeated the determinations at nor-
mal pressure with greater care than before.

Further work followed, devoted to determining the equilibrium at nor-

mal pressure and at 30 atmospheres over an extended range of temperatures,
to calculating the heat of formation of ammonia from the elements at nor-
mal temperature and at the threshold of red heat, and finally to obtaining
knowledge of its specific heat at increased temperature. (See Annotation on
p.340.)

During the course of these investigations, together with my young friend
and co-worker Robert le Rossignol, whose work I would like to mention

here with particular sincerity and gratitude, I took up once again, in 1908,

A M M O N I A S Y N T H E S I S F R O M E L E M E N T S

337

the problem of ammonia synthesis abandoned three years earlier. Immedi-
ately prior to this I

had become acquainted with the technical processes in the

liquefaction of air, and had simultaneously caught a glimpse of the formate

industry, which caused flowing carbon monoxide to act upon alkali under

heat and increased pressure, and I no longer considered it impossible to pro-
duce ammonia on a technical scale under high pressure and at high temper-
ature. But the unfavourable opinion of colleagues taught me that an impres-
sive advance would be needed to arouse technical interest in the subject.

To begin with, it was clear that a change to the use of maximum pressure

would be advantageous. It would improve the point of equilibrium and

probably the rate of reaction as well. The compressor which we then pos-

sessed allowed gas to be compressed to 200 atmospheres, and thus deter-
mined our working pressure which could not easily be exceeded for any

very large series of experiments. In the neighbourhood of this pressure, the

catalysts, with which we had become familiar in the course of our equilib-
rium determinations, very easily provided a rapid combination of nitrogen
and hydrogen at above 700

C; this applied notably to manganese, followed

by iron.

To achieve impressive results, however, we needed to discover catalysts

which would induce rapid conversion at between 500

and 600

C. We hit

upon the idea of searching the sixth, seventh and eighth groups in the Peri-
odic System, whose principal metals chromium, manganese, iron and nickel
possessed very definite catalytic properties, for metals which acted even
more favourably; these we found in uranium and osmium. At the same time
we discovered in osmium an excellent example of the extent to which the
performance of a catalyst depends on its composition. When used at 200
atmospheres, both requirements which we deemed necessary to a techni-

cally-convincing conduct of the experiment, were met; the first concerned
the ammonia content, the second the amount of ammonia produced per
cubic centimetre of contact space per hour.

With a content of about 5% the circulation process described in

1905

was no longer a description of a method of synthesis, but a means of manu-
facture. With a yield of several grams of ammonia per hour per cubic centi-
metre of heated high-pressure chamber the dimensions of the chamber
could be made so small that we felt the objections from the industry must
d i s a p p e a r .

Finally we needed an improvement in the circulation system which could

act as model for technical realization; separating the synthesis of ammonia

338

1918 F.HABER

and its removal from the flow of gas by means of reducing the pressure was

not a suitable method. The cycle of ammonia production and removal must

clearly be achieved by the simplest possible means at a constant high pres-
sure. It seemed essential that the heat produced during the synthesis of am-
monia should be removed from exhaust gases, where it had only a deleteri-
ous effect, and be led to the fresh gas so that the process itself yielded the heat
required for its operation. The construction and operation (carried out in
collaboration with Robert le Rossignol) of a small-scale plant which suited
these requirements, together with the performance of the new catalysts men-

tioned, was indeed sufficient to persuade the "Badische Anilin- und Soda-
fabrik" which thus far had devoted its attention to the indirect method of
producing ammonia by means of the nitrides of aluminium, silicium and ti-

tanium, to undertake high-pressure synthesis from the elements.

The company then studied the catalysts on a large scale, using far more

substantial means, and discovered ways, in the temperature employed in
their production plant and particularly in the deliberate use of inert ma-
terials, of improving the performance of poor catalysts to the level of os-
mium and uranium. Their results were, indeed, important in the case of the
classic ammonia catalyst employed by Ramsay and Young, namely iron.
They also discovered an improvement in the design of the oven which over-
came the effect of hydrogen on the carbon content of steel which they had
observed over a long period of operation.

The main work of the company however, was in substituting electrolytic

hydrogen, with which we conducted our experiments, for water-gas hy-
drogen which introduced impurities. The difficulties encountered by the
Technical Director Dr. Bosch resembled those which his predecessor Knietsch
had overcome with equal success in the course of his technical application of
the sulphuric acid contact process. Dr. Bosch has made a large-scale industry

of ammonia synthesis.

Present-day industrial working pressures in the vicinity of 200 atmo-

spheres, a working temperature of about 500-600

C, circulation under

constant high pressure, and the method of heat exchange from exhaust to

inlet gas are all main features of laboratory work which have been retained.

Recently Claude has announced an improvement of the process in the ap-

plication of higher pressures. The pressure range around 200 atmospheres
was originally chosen since it represented the limit of easily attainable levels
at the current stage of development in compressor technique. In subsequent

experiments Mr. Greenwood and I have gone as far as 370 atmospheres. An

A M M O N I A S Y N T H E S I S F R O M E L E M E N T S

339

increase in pressure is basically only of interest if it considerably reduces the

temperature of rapid conversion without creating fresh technical difficulties.

From the tabulated equilibria (p. 336), it can be seen that the change from

normal pressure to 200 atmospheres creates favourable equilibrium condi-
tions - existing between 200

and 300

C - at a temperature 300

C higher,

which stimulates more greatly the activity of the catalysts. Why a higher
temperature is needed is a question which we must leave to a more enlight-
ened period of science to answer. The heterogenous catalysis of the gas re-

actions is a process which in the initial phase apparently represents an elec-

trodynamic distortion of the molecule by the atomic fields at the boundary
of the solid catalyst material with the gas space; it is thus a phenomenon
from a field of molecular physics into which Stark’s discovery had just given

us a first glimpse.

The synthesis of ammonia from the elements is a result which physical

chemistry was bound to reach. The notion of the reversibility of the break-
down of ammonia was already held by Deville, Ramsay and Young, and by

1901 Le Chatelier had already given thought to the effects of temperature

and pressure. Failure of the first attempts at synthesis however led him to
abandon the matter and to publish his deliberations only in the obscurity of
a French patent taken out under a foreign name. This only came to my no-
tice a long time after the successful conclusion of my experiments.

The solution to the problem which has been found assumes its importance

from the fact that very high temperature levels are not employed and that
this makes the ratio of coal consumption to nitrogen production more fa-
vourable than is the case with other processes. Results are enough to show
that, in combination with other methods of nitrogen fixation which

have

mentioned, they relieve us of future worries caused by the exhaustion of the
saltpetre deposits that has threatened us these 20 years.

It may be that this solution is not the final one. Nitrogen bacteria teach us

that Nature, with her sophisticated forms of the chemistry of living matter,
still understands and utilizes methods which we do not as yet know how to

imitate. Let it suffice that in the meantime improved nitrogen fertilization
of the soil brings new nutritive riches to mankind and that the chemical
industry comes to the aid of the farmer who, in the good earth, changes
stones into bread.

340

1918 F.HABER

Annotation to p.336

The results, in brief, were as follows:
(a) Actual specific heat Cp of the ammonia gas per mol at constant pressure between

309

C and 523

C is:

Cp = 8.62 + 3.5 x 10

-3t

+ 5.1 x 10

-6

(b) Heat of formation Q of the ammonia gas at constant pressure in gramcalories per

mol from the elements at t

C is:

Q =

10,950

+ 4.85t - 0.93

1 0

-3

-1.7 X 10

-6

(3 Vol. H

+ 1 Vol. N

The following expression has been used for the calculation:

Also expressions with higher temperature links may be adapted to the observations. A

rational expression can only then be postulated when a rational statement concerning

the specific heat of all three participant gases has succeeded.